Search

Erratum to: FragClust and TestClust, two informatics tools for chemical structure hierarchical clustering analysis applied to lipidomics. The example of Alzheimer’s disease

Background and aims Obesity is predictive of metabolic syndrome (metS), type 2 diabetes, cardiovascular (CV) disease and cancer. The aim of the study is to assess the risk of incident cancer connected to obesity and metS in a Mediterranean population characterized by a high prevalence of obesity. Methods and results As many as 1133 subjects were enrolled in two phases and followed for 25 years (859 subjects) or 11 years (274 subjects) and incident cancer was registered in the follow-up period. Anthropometric measures and biochemical parameters were filed at baseline and evaluated as predictors of incident cancer by measuring hazards ratios (HR) using multivariate Cox parametric hazards models. Best predictive threshold for metabolic parameters and metS criteria were recalculated by ROC analysis. Fasting Blood Glucose >5.19 mmol/L [HR = 1.58 (1.0–2.4)] and the TG/HDL ratio (log 10 ) (Males > 0.225, Females > 0.272) [HR = 2.44 (1.3–4.4)] resulted independent predictors of survival free of cancer with a clear additive effect together with age classes [45–65 years, HR = 2.47 (1.3–4.4), 65–75 years HR = 3.80 (2.0–7.1)] and male gender [HR = 2.07 (2.3–3.1)]. Conclusions Metabolic disturbances are predictive of cancer in a 25 years follow-up of a Mediterranean population following a traditional Mediterranean diet. The high prevalence of obesity and metS and the observed underlying condition of insulin resistance expose this population to an increased risk of cardiovascular disease and cancer despite the healthy nutritional habits.

Lipidomic analysis is able to measure simultaneously thousands of compounds belonging to a few lipid classes. In each lipid class, compounds differ only by the acyl radical, ranging between C10:0 (capric acid) and C24:0 (lignoceric acid). Although some metabolites have a peculiar pathological role, more often compounds belonging to a single lipid class exert the same biological effect. Here, we present a lipidomics workflow that extracts the tandem mass spectrometry data from individual files and uses them to group compounds into structurally homogeneous clusters by chemical structure hierarchical clustering analysis (CHCA). The case-to-control peak area ratios of the metabolites are then analyzed within clusters. We created two freely available applications to assist the workflow: FragClust to generate the tables to be subjected to CHCA, and TestClust to perform statistical analysis on clustered data. We used the lipidomics data from the plasma of Alzheimer's disease (AD) patients in comparison with healthy controls to test the workflow. To date, the search for plasma biomarkers in AD has not provided reliable results. This article shows that the workflow is helpful to understand the behavior of whole lipid classes in plasma of AD patients. [Figure not available: see fulltext.]

Background Severe hypertriglyceridemia (HTG) may result from mutations in genes affecting the intravascular lipolysis of triglyceride (TG)-rich lipoproteins. Objective The aim of this study was to develop a targeted next-generation sequencing panel for the molecular diagnosis of disorders characterized by severe HTG. Methods We developed a targeted customized panel for next-generation sequencing Ion Torrent Personal Genome Machine to capture the coding exons and intron/exon boundaries of 18 genes affecting the main pathways of TG synthesis and metabolism. We sequenced 11 samples of patients with severe HTG (TG>885mg/dL–10mmol/L): 4 positive controls in whom pathogenic mutations had previously been identified by Sanger sequencing and 7 patients in whom the molecular defect was still unknown. Results The customized panel was accurate, and it allowed to confirm genetic variants previously identified in all positive controls with primary severe HTG. Only 1 patient of 7 with HTG was found to be carrier of a homozygous pathogenic mutation of the third novel mutation of LMF1 gene (c.1380C>G–p.Y460X). The clinical and molecular familial cascade screening allowed the identification of 2 additional affected siblings and 7 heterozygous carriers of the mutation. Conclusions We showed that our targeted resequencing approach for genetic diagnosis of severe HTG appears to be accurate, less time consuming, and more economical compared with traditional Sanger resequencing. The identification of pathogenic mutations in candidate genes remains challenging and clinical resequencing should mainly intended for patients with strong clinical criteria for monogenic severe HTG.